INVESTIGATING THE EFFICACY OF REDUCED POINT DENSITY LIDAR TO MEASURE TOPOGRAPHICAL CHANGE IN COASTAL DUNES (FIRE ISLAND, NY)

The efficacy of LiDAR to detect topographical change within a beach-dune system is examined empirically with regard to scale. Small dimensional changes are beyond the capability of LiDAR to measure reliably. However, it can capture elevation well enough to track trends in topographical development and quantify storm impacts. The basis for this hypothesis is provided by the comparison of high accuracy GPS topographical survey data collected in June 2014 on Fire Island National Seashore and a DEM created from LiDAR concurrently collected by the USGS EAARL-B sensor, and then exploring the implications of the findings for geotemporal change analyses. Centimeter accurate GPS survey data was utilized to represent the true ground condition of the beach, foredune, and swale behind the dune during the LiDAR survey. This comparison provides a measure of error for this LiDAR survey conducted in a coastal environment and can be used to assess significance for metrics of dimensional change within it.

Challenges associated with representing dunal topography with a LiDAR derived DEM include the ground condition being misrepresented in a bare earth processed dataset as a result of dense vegetation. Non-ground data points left in a dataset can be reduced by selecting and retaining only the lowest point for given area windows in a LiDAR point cloud. As a result, the point density is reduced and, ideally, only points that represent a laser pulse that has been returned from the ground surface remain. This method, which can be applied consistently to any LiDAR dataset, can help minimize the difference between the resulting DEM and the true ground surface. Additionally, horizontal error present in the dataset can produce larger differences in elevation from the true ground surface in high slope areas, such as dunes, than those found on flat surfaces. This experiment serves as an assessment of the composite error for all processes involved in creating a LiDAR DEM, which includes the aforementioned sources, as well as others associated with the data collection or DEM interpolation. It is also likely that there is greater uncertainty associated with a DEM's representation of the surface within some geomorphological features than in others, as the topography that typifies a feature often lends to characteristics that compromise the DEM’s accuracy.